Inflect v2: Two ultra-tiny TTS models with 4M and 10M parameters
Inflect v2 introduces ultra-tiny TTS models with 4M and 10M parameters, enhancing usability.
Over the past month, I explored the threshold where extremely small TTS models transition from being size experiments to genuinely useful tools. Today, I am releasing Inflect v2, featuring two local text-to-speech models: Inflect-Nano-v2 with 3.96M parameters and Inflect-Micro-v2 with 9.36M parameters. These models include all necessary components for text processing and speech generation, allowing them to run locally without external dependencies.
This synthesis was produced from its source by AI; there is no human editor or manual review step. How we work